| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.8, | |
| "global_step": 3000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 1.6800000000000002e-05, | |
| "loss": 0.5178, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.946119307248236, | |
| "eval_f1": 0.9269931515699703, | |
| "eval_loss": 0.20892712473869324, | |
| "eval_precision": 0.9278323848939473, | |
| "eval_recall": 0.9261554350632585, | |
| "eval_runtime": 2.7187, | |
| "eval_samples_per_second": 367.826, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 1.3600000000000002e-05, | |
| "loss": 0.2064, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9685695958948044, | |
| "eval_f1": 0.9600723607701254, | |
| "eval_loss": 0.1718911975622177, | |
| "eval_precision": 0.9609415416451113, | |
| "eval_recall": 0.9592047508391428, | |
| "eval_runtime": 2.6825, | |
| "eval_samples_per_second": 372.791, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 1.04e-05, | |
| "loss": 0.1378, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9693713919178961, | |
| "eval_f1": 0.961869302863012, | |
| "eval_loss": 0.16582955420017242, | |
| "eval_precision": 0.95658835546476, | |
| "eval_recall": 0.9672088820036148, | |
| "eval_runtime": 2.6697, | |
| "eval_samples_per_second": 374.576, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 7.2000000000000005e-06, | |
| "loss": 0.0884, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.0616, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9757857601026299, | |
| "eval_f1": 0.9698453608247423, | |
| "eval_loss": 0.1476062387228012, | |
| "eval_precision": 0.9680987908412657, | |
| "eval_recall": 0.9715982442550994, | |
| "eval_runtime": 4.3214, | |
| "eval_samples_per_second": 231.405, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 8.000000000000001e-07, | |
| "loss": 0.0426, | |
| "step": 3000 | |
| } | |
| ], | |
| "max_steps": 3125, | |
| "num_train_epochs": 5, | |
| "total_flos": 228175031247744.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |